Analysis
Product Requirement Ambiguity Detection
Given a product requirement, identify undefined actors, missing states, conflicting constraints, untestable language, and unanswered acceptance questions.
Engineering & ProductPer requirement
Not benchmarked
No result on this tier
Token profile pending
— /100K
Run this tier to add it
—
Granite 4.1 8B
8B · Off-the-shelf
230 in · 974 out · 14.25s p50
$10.89 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/4/2026
OpenRouter · $0.0500 input · $0.1000 output /1M
TrustedRouter · $0.0527 input · $0.1055 output /1M
92%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
225 in · 1,093 out · 10.64s p50
$110 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.3376 input · $0.9390 output /1M
OpenRouter · $0.2900 input · $1.14 output /1M
99%
Best accuracy
GPT-5.4
Hosted · Off-the-shelf
226 in · 2,000 out · 20.42s p50
$3057 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $2.50 input · $15.00 output /1M
TrustedRouter · $2.64 input · $15.82 output /1M
25%
All 12 model results and methodology
Generated 97/100 examples — partial set accepted at the success threshold
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 82% | 97/97 | $40.26 |
| Qwen3 235B A22B | mid | 92% | 97/97 | $257 |
| DeepSeek V3 | mid | 99% | 97/97 | $110 |
| Mistral Large 2407 | mid | 94% | 97/97 | $130 |
| Arcee Trinity Large Thinking | mid | 98% | 97/97 | $130 |
| GPT-5.4 | frontier | 25% | 97/97 | $3057 |
| Claude Opus 4.7 | frontier | 18% | 97/97 | $5191 |
| Gemini 3.1 Pro Preview | frontier | 18% | 97/97 | $2423 |
| Gemma 4 E4B IT | small | 89% | 97/97 | $12.17 |
| Granite 4.1 8B | small | 92% | 97/97 | $10.89 |
| Ministral 8B Instruct 2410 | small | 63% | 97/97 | $15.21 |
| Qwen3 4B Instruct 2507 | small | 71% | 97/97 | $120 |
LLM-judge pass rate on 97 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-10T08:54:53.273Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock